Critique Questions Provenge Benefit, Prompts Ire

Rebuttal to Be Published Next Month

Nick Mulcahy

April 13, 2012

April 13, 2012 — A controversial commentary that questions whether the novel immunotherapy sipuleucel-T (Provenge, Dendreon) has any benefit in patients with advanced prostate cancer has provoked swift criticism from the manufacturer, the agent's investigators, and at least one expert.

The essay, published in the February 22 issue of the Journal of National Cancer Institute, puts forth a dramatic hypothesis: the 4.1-month overall survival benefit seen in men treated with sipuleucel-T in its pivotal phase 3 trial might be the result of a harmful effect of the "placebo intervention," rather than a beneficial effect of the vaccine.

Lead author Marie Huber, MPhil, who graduated from Cambridge University in the United Kingdom, became "fascinated" with sipuleucel-T when she worked as a hedge fund healthcare analyst in New York City. She left that job to avoid any conflict of interest.

Her coauthors are 2 European academic urologists, both of whom report being advisors to Dendreon in the past, and an American immunologist.

A prominent sipuleucel-T researcher castigated the essay.

It has no scientific basis.

"Many were surprised that this commentary was published since it has no scientific basis," Philip Kantoff, MD, told Medscape Medical News in an email. Dr. Kantoff was a principal investigator of the pivotal trial — the Immunotherapy for Prostate Adenocarcinoma Treatment (IMPACT) study — and is chief clinical research officer at the Dana-Farber Cancer Institute in Boston, Massachusetts.

The main arguments in the essay were "dismissed 2 years ago" by the US Food and Drug Administration (FDA) and the Center Medicare and Medicaid Services (CMS), he said.

A Dendreon official emphasized the authority of the federal reviews in the United States.

Sipuleucel-T "has gone through an incredibly rigorous review process," said Mark Frohlich, MD, chief medical officer of Dendreon in Seattle, Washington. That process has included 2 FDA reviews, which led to the vaccine's approval for the treatment of asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer (mCRPC) in 2010, and the CMS coverage decision for sipuleucel-T, which included an extra efficacy and safety analysis that the agency does not typically perform.

Dr. Frohlich, Dr. Kantoff, and other investigators are currently working on an official response to the essay. It is anticipated that it and other rebuttals will be published in the May issue of the journal.

In the meantime, the essayists are in the spotlight, having concluded that "a detrimental effect of the placebo intervention is at least as plausible as a beneficial effect of sipuleucel-T as an explanation of the survival difference observed in the IMPACT trial."

They acknowledge that their theory is not proven and should be seen as an "alternative explanation of the IMPACT trial outcome."

"Over time, I think this will be recognized as a spurious argument," said Dr. Frohlich in an interview with Medscape Medical News. "It would be unfortunate if patients were dissuaded from using the product by not-valid arguments."

Was the Placebo Harmful?

Dendreon and the IMPACT investigators have had their study results questioned in the past.

When the final results from IMPACT were published (N Engl J Med. 2010;363:411-422), Dan Longo, MD, a former cancer immunologist at the National Institutes of Health who is now deputy editor of the New England Journal of Medicine, noted that it was "surprising" that the prolongation of survival in the trial occurred "without a measurable antitumor effect."

In an editorial that accompanied the IMPACT study (N Engl J Med. 2010;363: 479-481), Dr. Longo said that sipuleucel-T was designed to stimulate a "host antitumor response" but that a lack of a related measurable effect "raises concern that the results could have been influenced by an unmeasured prognostic variable that was accidentally imbalanced in study-group assignments."

This lack has also been seen with another experimental prostate cancer vaccine — PROSTVAC-VF. Dr. Kantoff and colleagues noted a discordance between overall survival and tumor response in a phase 2 trial with this product, and described the phenomena as a possible "emerging theme" with immunotherapies (J Clin Oncol. 2010;28:1099-1105).

It is in this context of uncertainty about efficacy, write Huber and her coauthors, that they reviewed the IMPACT design and data.

The IMPACT trial design required that all patients, whether in the drug or placebo group, undergo apheresis and have peripheral-blood mononuclear cells harvested from their blood.

It was an active intervention.

In the treatment group, all of the cells were then immunologically manipulated and returned to the patients.

In the placebo group, the cells were not manipulated, but only one third of the cells were returned to the patients. Two thirds were frozen and kept in reserve. The reason for holding back the majority of the placebo patients' cells was to facilitate their receiving the vaccine in the event of the need for salvage therapy.

This study design had no ill intent, but the placebo used in IMPACT is not the equivalent of a sugar pill, said Huber in an interview with Medscape Medical News.

"It was an active intervention," she said about the placebo, adding that it could have had a detrimental effect on immune systems.

In an interview with Medscape Medical News, Dr. Frohlich countered that an analysis by the IMPACT investigators fully addressed this matter (J Urol. 2011;186:877-881). "There was no evidence that leukapheresis led to immunodepletion," write the investigators, led by Simon Hall, MD, from the Mount Sinai School of Medicine in New York City. The study also reported a normal measured cell count after the treatment and placebo processes were complete. "Lymphocytes removed from the peripheral circulation are rapidly replenished by regeneration and redistribution," explain the authors.

It is "highly unlikely" that the process of leukapheresis would affect the men's immune systems, explained Dr. Frohlich, because each of the 3 apheresis procedures used in the study removed only a tiny fraction of the total pool of blood cells.

He also pointed out that the adverse events data from IMPACT do not support the idea that the placebo was immunologically injurious. For instance, there was no difference in infections or lymphocyte counts between the treatment and placebo groups in the trial, said Dr. Frohlich.

In a review of sipuleucel-T and other immunotherapies for prostate cancer, a group of European urologists agreed that there was likely no negative impact on the immune system from the placebo (Eur Urol. 2012;61:290-306).

"Only a small number of total circulating cells are removed by leukapheresis," write those authors, led by Steven Joniau, MD, from the Department of Urology, University Hospital, in Leuven, Belgium.

The process of removing and returning blood cells to the body is well established, they explain. "For decades, similar procedures have been performed in healthy subjects and in cancer patients with no demonstrable harm," they add.

Dr. Joniau and his colleagues endorse sipuleucel-T. IMPACT provides "high-level evidence in favor of vaccination with the stimulated leukapheresis product," they note.

They also give some credence to the assertions by Huber and colleagues — only to dismiss them as "unlikely."

"The [IMPACT] trial...raises doubts about the appropriateness of a placebo arm in future trials with similar patients," they write.

"An unfavorable effect in the control group could result from the leukapheresis procedure itself; however, this procedure is unlikely to result in the treatment group's survival advantage," they continue, suggesting that the 4.1-month difference is too big to be accounted for by the procedure.

Cells Handled Differently in Study Groups

Huber and colleagues say that the unique trial design of IMPACT might have resulted in a degeneration of the cells that were ultimately returned to the placebo patients. That could have happened because the placebo patients' cells that were returned were handled differently than the cells of patients in the treatment group.

The cells taken from patients in the treatment group were incubated at 37 °C for 36 to 44 hours.

However, the cells taken from patients in the placebo group were stored at a much lower temperature — 2 to 8 °C. "In our experience, storage of isolated mononuclear cells at 2 to 8 °C for 36 to 44 hours can result in the death of most, if not all, of those cells," Huber and colleagues write.

In addition, the cells from the patients in the 2 study groups were processed in different mediums; the ramifications of this are not known, they note.

Cells processed in sipuleucel-T were incubated with the granulocyte-macrophage colony-stimulating factor (GM-CSF)/prostatic acid phosphatase (PAP) chimeric protein, whereas cells processed in placebo were stored in a medium containing no GM-CSF.

A cancer immunologist contacted by Medscape Medical News for comment suggested that this difference is not ideal.

"Scientifically, it would have been best if placebo patients were reinfused with the 'processed cells' that were incubated with GM-CSF, following the same procedure as for sipuleucel-T-treated patients. In this way, one can exactly determine the net effect of the active component in the treatment [the PAP chimeric protein]," said Sjoerd H. van der Burg, PhD, professor of experimental cancer immunology and therapy at the Leiden University Medical Center in the Netherlands. This point was also made by Dr. Longo in his editorial.

For placebo, the goal was not to activate the cells.

Dr. Frohlich explained that these differences in temperature and storage mediums were purposeful. They were necessary to avoid activating cells from the placebo patients and helped preserve the placebo effect. "For placebo, the goal was not to activate the cells, but to return them unchanged. Adding GM-CSF, warming the cells, etc., would lead to activation of some of the cells and therefore may not have been a true placebo," he argued.

He also said that the company and its investigators, in cooperation with the FDA, tried to create the "best possible placebo," but that it was "not possible to create a product that was identical to [sipuleucel-T but inactive]."

Dr. van der Burg also said that devising a placebo can be challenging. "It is always difficult to think about the most appropriate placebo, as one does not want to harm patients in the control group," he noted.

Observations Prompt Questions and Responses

Huber and colleagues make a number of observations about the phase 3 clinical trial data on sipuleucel-T that led them to question its efficacy.

For instance, they observe that the benefit seen with sipuleucel-T appears to be limited to patients who are 65 years or older.

Using data taken from Dendreon's submission to the FDA, the IMPACT investigators report that among patients younger than 65 years, sipuleucel-T treatment appeared to have no effect on survival (hazard ratio of death, 1.41; 95% confidence interval [CI], 0.87 to 2.29). However, in patients 65 years or older, there was an effect on survival (hazard ratio of death, 0.58; 95% CI, 0.43 to 0.76).

"This observation suggests that the overall results were driven entirely by the differential survival in older patients," Huber and coauthors note. Sipuleucel-T seems to mostly benefit older patients, which is "counterintuitive," they say, because immunotherapies are "consistently less effective in the elderly." The finding raises "further questions" about how sipuleucel-T works, they say.

Dr. Frohlich objected to the claim that sipuleucel-T does not work in younger patients.

The published IMPACT data demonstrate that sipuleucel-T works in both older and younger patients, he said.

The results "indicate a positive treatment effect for [sipuleucel-T] for patients above and below the median age of 71," he said.

Age 71 — not age 65 — was chosen by the IMPACT investigators as the point of analysis because it was the median age of the study participants, Dr. Frohlich pointed out.

The FDA also rejected this concern about the effect of age on overall survival. An analysis of the pooled data from IMPACT and 2 smaller trials of sipuleucel-T indicated, in the opinion of the FDA reviewer, that "the subgroup of subjects who were less than 65 years of age also benefit from treatment with sipuleucel-T."

"The most striking observation from the new data was an unexpected 11-month difference in median survival between placebo patients younger than age 65 years and patients older than age 65 years (28.2 vs 17.2 months, respectively)," Huber and colleagues write.

This is important because age usually does not affect survival in patients with mCRPC who are treated with chemotherapy, they note. In other words, age has not been a factor that predicted patients' outcomes in this setting among chemotherapy-treated patients.

In IMPACT, age seems to be prognostic, said Huber. What could explain this? It is possible that the older men on placebo were vulnerable to any related detrimental impact from the placebo process and died more quickly, she said. The younger men might have been more robust and thus more capable of recovering.

But a prostate cancer expert suggested that this survival finding among men on placebo, which is from a post hoc subgroup analysis, is not strong evidence.

"After stating that 'post hoc subgroup analyses should be interpreted with caution,' the authors proceeded to perform such an analysis of the IMPACT trial," writes Carl Olsson, MD, in the March issue of the AUA News, a newsletter from the American Urological Association. He is chair of urology emeritus at Columbia University Medical Center in New York City.

Dr. Olsson also takes a swipe at Huber's qualifications in his piece, at the same time acknowledging the value of collaboration among specialists such as her coauthors (2 urologists and an immunologist).

"We are all used to the values of interdisciplinary conferences, which usually combine the experiential qualities of urology, clinical oncology, immunology, and other disciplines. However, 'healthcare analyst' and 'investment management' are talents that appear unseemly in major publications," he writes.

This is out of my comfort zone.

Huber, who has undergraduate degree in biochemistry and a master's degree from a program that combines medical science and business, said that she never planned to become an activist. "I am quite an introvert. This is out of my comfort zone," she said.

She hopes to return to her former field of employment soon; "I need to find myself an income again shortly," she said.

Huber has disclosed no relevant financial relationships. Coauthor Peter Iversen, MD, from the Department of Urology, University of Copenhagen, Rigshospitalet, in Denmark, reports advisory and/or lecturing relationships with Dendreon, Bavarian Nordic, AstraZeneca, Ferring, Amgen, Astellas, Janssen, and sanofi-aventis. Coauthor Chris Parker, MD, from The Royal Marsden NHS Foundation Trust, United Kingdom, reports advisory and/or lecturing relationships with Dendreon, Takeda, Janssen, Bayer, and Amgen. Dr. Kantoff reports being a consultant to Dendreon, BN-IT, Bayer, Janssen, Bellicum, and BIND. Dr. van der Burg reports being a member of the strategic advisory boards of ISA Pharmaceuticals, Transgene SA, Dandrit Biotech, and DCprime.

J Natl Cancer Inst. 2012;104:273-279. Abstract